Skip to content
SPECIFICTION
Get listed

RunbookAI

userunbook.ai

Linux
Visit userunbook.ai
Spec score 6.8 #17 of 24 in Runbook Automation Software

What it is

RunbookAI is a self-hosted, MIT-licensed incident-response tool for investigating production problems. Its workflow collects incident context, ranks hypotheses, checks them against infrastructure, identifies a root cause, and recommends remediation. Users can ask natural-language questions across AWS, Kubernetes, and CloudWatch, then execute step-by-step runbooks. Queries are read-only by default; every mutating action needs explicit human approval. Investigations record hypotheses, evidence, proposed actions, approval decisions, and execution results in a local audit trail. RunbookAI indexes knowledge from local files, Confluence, and Google Drive. Documented integrations include AWS, Kubernetes, PagerDuty, OpsGenie, Slack, and Claude Code. A self-hosted shared knowledge server exposes runbooks, postmortems, and known issues through REST and an MCP endpoint for Claude Code. CLI mode is available, with no hosted service or control plane. The maker says it runs inside your infrastructure, sends no telemetry, and sends model requests directly to the configured provider without proxying or storing prompts. Requirements include Bun 1.0+ or Node.js 20+ and an Anthropic API key for Claude.

Spec sheet

Purpose
RunbookAI investigates production incidents by forming hypotheses, gathering evidence and recommending fixes.Read 4 Oct 2026 · userunbook.ai
Intended users
The maker describes it as an incident-response tool for SRE teams and is onboarding teams running production workloads on AWS and Kubernetes.Read 4 Oct 2026 · userunbook.ai
Deployment
RunbookAI offers CLI mode and a self-hosted shared knowledge server, with no hosted service or control plane.Read 4 Oct 2026 · userunbook.ai
Integrations
Documented integrations include AWS, Kubernetes, PagerDuty, OpsGenie, Slack and Claude Code.Read 4 Oct 2026 · userunbook.ai
Support
The documentation directs users with questions or issues to GitHub Issues or Discussions.Read 4 Oct 2026 · userunbook.ai
Audit trail
RunbookAI logs investigation hypotheses, evidence, proposed actions, approval decisions and execution results locally.Read 4 Oct 2026 · userunbook.ai
Claude Code
The shared knowledge server exposes an MCP endpoint that Claude Code can use to query operational knowledge.Read 4 Oct 2026 · userunbook.ai
Data handling
The maker says RunbookAI runs within the user's infrastructure, sends no telemetry, and sends LLM calls directly to the configured provider without proxying or storing prompts.Read 4 Oct 2026 · userunbook.ai
Deployment and license
RunbookAI is self-hosted and released under the MIT License.Read 4 Oct 2026 · userunbook.ai
Incident workflow
Its investigation workflow gathers incident context, ranks hypotheses, tests them against infrastructure, identifies a root cause and suggests remediation.Read 4 Oct 2026 · userunbook.ai
Infrastructure queries
The CLI supports natural-language queries across AWS, Kubernetes, and a knowledge base.Read 4 Oct 2026 · userunbook.ai
Investigation
Its investigation workflow gathers incident context, tests ranked hypotheses against infrastructure, and suggests remediation steps.Read 4 Oct 2026 · userunbook.ai
Knowledge sources
RunbookAI can index knowledge from local files, Confluence and Google Drive.Read 4 Oct 2026 · userunbook.ai
Open source
RunbookAI is released under the MIT License.Read 4 Oct 2026 · userunbook.ai
Requirements
The documentation lists Bun 1.0+ or Node.js 20+ and an Anthropic API key for Claude as requirements.Read 4 Oct 2026 · userunbook.ai
Runbooks
It can execute step-by-step runbooks, with approval gates for mutating actions.Read 4 Oct 2026 · userunbook.ai
Safety
Infrastructure queries are read-only by default, and every mutating action requires explicit human approval.Read 4 Oct 2026 · userunbook.ai
Safety controls
Every mutating action requires explicit human approval, while infrastructure queries are read-only by default.Read 4 Oct 2026 · userunbook.ai
Shared knowledge server
A self-hosted server lets a team query shared runbooks, postmortems and known issues through a REST API and an MCP endpoint.Read 4 Oct 2026 · userunbook.ai

Strong points and limits

Strong points

  • Self-hosted MIT-licensed deployment
  • Read-only defaults and approval gates
  • Local investigation audit trail
  • Shared REST and MCP knowledge server
  • AWS and Kubernetes integrations

Limits

  • Requires Bun or Node.js
  • Requires an Anthropic API key
  • No hosted service or control plane

Who it suits

RunbookAI suits SRE teams operating production workloads on AWS and Kubernetes. It is intended for organizations that can self-host and provide the listed runtime and Claude requirements.

Verdict

RunbookAI emphasizes controlled incident investigation, operational knowledge, and explicit approval for changes. Its self-hosted design and runtime requirements should fit teams managing their own infrastructure.

Where Specifiction places it

Alternatives to RunbookAIAll alternatives →